Keepalived 的script及双主
Posted
tags:
篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了Keepalived 的script及双主相关的知识，希望对你有一定的参考价值。
环境：
K1：192.168.11.26
K2：192.168.11.28
VIP1：192.168.11.17
VIP2：192.168.11.18
一、vrrp_script
1、K1配置
[[email protected] keepalived]# cat keepalived.conf #配置内容 ! Configuration File for keepalived global_defs { notification_email { [email protected] } notification_email_from [email protected] smtp_server 127.0.0.1 smtp_connect_timeout 30 router_id LVS_DEVEL # vrrp_mcast_group4 224.0.100.18 } vrrp_script chk_down { #定义vrrp_script脚本扩展 script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0" #判断文件是否存在，也可以定义脚本 interval 3 #检查周期 weight -10 #文件存在权重-10 根据上面script定义 } vrrp_instance VI_1 { state MASTER interface eth0 virtual_router_id 51 priority 100 advert_int 1 authentication { auth_type PASS auth_pass 6Nb6zjWB } virtual_ipaddress { 192.168.11.17 } track_script { #跟踪script，如果不写，则不调用也就是不会执行我们是定义的vrrp_script定义的内容 chk_down } notify_master "/etc/keepalived/notify.sh master" notify_backup "/etc/keepalived/notify.sh backup" notify_fault "/etc/keepalived/notify.sh fault" }
2、K2配置
[[email protected] keepalived]# cat keepalived.conf #配置内容 ! Configuration File for keepalived global_defs { notification_email { [email protected] } notification_email_from [email protected] smtp_server 127.0.0.1 smtp_connect_timeout 30 router_id LVS_DEVEL # vrrp_mcast_group4 224.0.100.18 } vrrp_script chk_down { #定义vrrp_script脚本扩展 script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0" #判断文件是否存在，也可以定义脚本 interval 3 #检查周期 weight -10 #文件存在权重-10 根据上面script定义 } } vrrp_instance VI_1 { state BAKCUP #填写BACKUP，因为K1是MASTER interface eth0 virtual_router_id 51 priority 98 #优先级也需要设定比K1小 advert_int 1 authentication { auth_type PASS auth_pass 6Nb6zjWB } virtual_ipaddress { 192.168.11.17 } track_script { #跟踪script，如果不写，则不调用也就是不会执行我们是定义的vrrp_script定义的内容 chk_down } notify_master "/etc/keepalived/notify.sh master" notify_backup "/etc/keepalived/notify.sh backup" notify_fault "/etc/keepalived/notify.sh fault" }
3、测试
#启动keepalived服务，这里就不演示了 [[email protected] keepalived]# touch down #在K1MASTER上touche down文件 [[email protected] keepalived]# ls #查看是否存在，发现有down文件 down keepalived.conf keepalived.conf.bak keepalived.conf.real notify.sh [[email protected] keepalived]# tail -f /var/log/messages #在另外一个K1的session tail -f Jun 24 17:43:57 web2 Keepalived_vrrp[9530]: VRRP_Script(chk_down) failed #发现我们定义的track_script（chk_down） failed ，自动权限-10 Jun 24 17:43:58 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Received higher prio advert #发现更高优先级的 Jun 24 17:43:58 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Entering BACKUP STATE #自己转为BACKUP STATE Jun 24 17:43:58 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) removing protocol VIPs. #迁移VIPs协议 Jun 24 17:43:58 web2 Keepalived_healthcheckers[9529]: Netlink reflector reports IP 192.168.11.17 removed #宣布删除IP 192.168.11.17 [[email protected] keepalived]# rm -f down #删除down文件 You have new mail in /var/spool/mail/root #查看，down的确被删除 [[email protected] keepalived]# ls keepalived.conf keepalived.conf.bak keepalived.conf.real notify.sh [[email protected] keepalived]# tail -f /var/log/messages #切换回刚刚的session Jun 24 17:43:57 web2 Keepalived_vrrp[9530]: VRRP_Script(chk_down) failed #发现我们定义的track_script（chk_down） failed ，自动权限-10，此时权限为90 Jun 24 17:43:58 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Received higher prio advert #发现更高优先级的 Jun 24 17:43:58 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Entering BACKUP STATE #自己转为BACKUP STATE Jun 24 17:43:58 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) removing protocol VIPs. #迁移VIPs协议 Jun 24 17:43:58 web2 Keepalived_healthcheckers[9529]: Netlink reflector reports IP 192.168.11.17 removed #宣布删除IP 192.168.11.17 ========================================================割了，实际是连在一起的，只不过为了方便解释，割了============================================== Jun 24 17:50:00 web2 Keepalived_vrrp[9530]: VRRP_Script(chk_down) succeeded #发现我们定义的tarck_script（chk_down） succeeded，此时权限为100，就是把之前-10的加回来了 Jun 24 17:50:00 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) forcing a new MASTER election #推选了一个新的MATER Jun 24 17:50:00 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) forcing a new MASTER election #推选了一个新的MASTER Jun 24 17:50:01 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Transition to MASTER STATE #正在进入MASTER状态中 Jun 24 17:50:02 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Entering MASTER STATE #进入MASTER状态 Jun 24 17:50:02 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) setting protocol VIPs. #设定VIPs协议 Jun 24 17:50:02 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 192.168.11.17 #从eth0发送192.168.11.17免费的ARPs Jun 24 17:50:02 web2 Keepalived_healthcheckers[9529]: Netlink reflector reports IP 192.168.11.17 added #添加192.168.11.17 Jun 24 17:50:07 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 192.168.11.17 #从eth0发送192.168.11.17免费的ARPs
4、通过vrrp_script简单检查服务并设定漂移
[[email protected] keepalived]# cat keepalived.conf #查看修改后的配置文件内容 ! Configuration File for keepalived global_defs { notification_email { [email protected] } notification_email_from [email protected] smtp_server 127.0.0.1 smtp_connect_timeout 30 router_id LVS_DEVEL # vrrp_mcast_group4 224.0.100.18 } vrrp_script chk_down { script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0" interval 3 weight -10 } vrrp_script chk_httpd { #增加一个新的vrrp_script，为chk_httpd script "killall -0 httpd" #killall -0 信号，如果服务存在返回0 interval 3 #检查周期 weight -10 #httpd服务存在权重-10 } vrrp_instance VI_1 { state BACKUP interface eth0 virtual_router_id 51 priority 98 advert_int 1 authentication { auth_type PASS auth_pass 6Nb6zjWB } virtual_ipaddress { 192.168.11.17 } track_script { chk_down chk_httpd #增加了一个新的track_script chk_httpd } notify_master "/etc/keepalived/notify.sh master" notify_backup "/etc/keepalived/notify.sh backup" notify_fault "/etc/keepalived/notify.sh fault" } [[email protected] keepalived]# service keepalived stop #在K2停止keepalived Stopping keepalived: [ OK ] [[email protected] keepalived]# netstat -tnlp| grep httpd #查看httpd服务是否启动 tcp 0 0 :::80 :::* LISTEN 19187/httpd [[email protected] keepalived]# service keepalived start #启动keepalived服务 Starting keepalived: [ OK ] [[email protected] keepalived]# ip add sh #查看IP，发现：inet 192.168.11.17/32 scope global eth0 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 00:0c:29:2d:ba:b0 brd ff:ff:ff:ff:ff:ff inet 192.168.11.28/24 brd 192.168.11.255 scope global eth0 inet 192.168.11.17/32 scope global eth0 inet6 fe80::20c:29ff:fe2d:bab0/64 scope link valid_lft forever preferred_lft forever [[email protected] keepalived]# service keepalived stop #在K1停止keepalived 正在启动 keepalived： [确定] [[email protected] keepalived]# netstat -tunlp | grep httpd #查看httpd是否启动 tcp 0 0 :::80 :::* LISTEN 5457/httpd [[email protected] keepalived]# service keepalived start #启动keepalived [[email protected] keepalived]# ip add sh #查看IP，发现：192.168.11.17被抢占过来了 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 00:0c:29:ad:98:5c brd ff:ff:ff:ff:ff:ff inet 192.168.11.26/24 brd 192.168.11.255 scope global eth0 inet 192.168.11.17/32 scope global eth0 inet6 fe80::20c:29ff:fead:985c/64 scope link valid_lft forever preferred_lft forever [[email protected] keepalived]# service httpd stop #在K1停止httpd服务 停止 httpd： [确定] You have new mail in /var/spool/mail/root [[email protected] keepalived]# ip add sh #查看IP，发信192.168.11.17不见了 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 00:0c:29:ad:98:5c brd ff:ff:ff:ff:ff:ff inet 192.168.11.26/24 brd 192.168.11.255 scope global eth0 inet6 fe80::20c:29ff:fead:985c/64 scope link valid_lft forever preferred_lft forever [[email protected] keepalived]# tail -f /var/log/messages #在K1停止httpd服务器前，开启的一个新session Jun 24 18:26:43 web2 Keepalived_vrrp[11701]: VRRP_Script(chk_httpd) failed #定义的trackscript（chk_httpd）faileld Jun 24 18:26:44 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Received higher prio advert #发现更高的优先级 Jun 24 18:26:44 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Entering BACKUP STATE #进入BACKUP状态 Jun 24 18:26:44 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) removing protocol VIPs. #迁移VIPs协议 Jun 24 18:26:44 web2 Keepalived_healthcheckers[11700]: Netlink reflector reports IP 192.168.11.17 removed #宣布删除IP 192.168.11.17 [[email protected] keepalived]# service httpd start #启动httpd服务 正在启动 httpd：httpd: apr_sockaddr_info_get() failed for K1 httpd: Could not reliably determine the server‘s fully qualified domain name, using 127.0.0.1 for ServerName [确定] You have new mail in /var/spool/mail/root [[email protected] keepalived]# ip add sh #查看IP，发现：192.168.11.17被抢占过来了 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 00:0c:29:ad:98:5c brd ff:ff:ff:ff:ff:ff inet 192.168.11.26/24 brd 192.168.11.255 scope global eth0 inet 192.168.11.17/32 scope global eth0 inet6 fe80::20c:29ff:fead:985c/64 scope link valid_lft forever preferred_lft forever [[email protected] keepalived]#tail -f /var/log/messages #在K1停止httpd服务器前，开启的一个新session Jun 24 18:26:43 web2 Keepalived_vrrp[11701]: VRRP_Script(chk_httpd) failed #定义的trackscript（chk_httpd）faileld Jun 24 18:26:44 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Received higher prio advert #发现更高的优先级 Jun 24 18:26:44 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Entering BACKUP STATE #进入BACKUP状态 Jun 24 18:26:44 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) removing protocol VIPs. #迁移VIPs协议 Jun 24 18:26:44 web2 Keepalived_healthcheckers[11700]: Netlink reflector reports IP 192.168.11.17 removed #宣布删除IP 192.168.11.17 ========================================================割了，实际是连在一起的，只不过为了方便解释，割了============================================== Jun 24 18:31:10 web2 Keepalived_vrrp[11701]: VRRP_Script(chk_httpd) succeeded #发现我们定义的tarck_script（chk_httpd） succeeded，此时权限为100，就是把之前-10的加回来了 Jun 24 18:31:10 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) forcing a new MASTER election #推选了一个新的MATER Jun 24 18:31:10 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) forcing a new MASTER election #推选了一个新的MATER Jun 24 18:31:11 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Transition to MASTER STATE #正在进入MASTER状态中 Jun 24 18:31:12 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Entering MASTER STATE #进入MASTER状态 Jun 24 18:31:12 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) setting protocol VIPs. #设定VIPs协议 Jun 24 18:31:12 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 192.168.11.17 #从eth0发送192.168.11.17免费的ARPs Jun 24 18:31:12 web2 Keepalived_healthcheckers[11700]: Netlink reflector reports IP 192.168.11.17 added #添加192.168.11.17 Jun 24 18:31:17 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 192.168.11.17 #从eth0发送192.168.11.17免费的ARPs
二、双主模型
[[email protected] keepalived]# cat keepalived.conf #查案配置内容，K1的就不贴上来了，新增段配置改为相反的 ! Configuration File for keepalived global_defs { notification_email { [email protected] } notification_email_from [email protected] smtp_server 127.0.0.1 smtp_connect_timeout 30 router_id LVS_DEVEL # vrrp_mcast_group4 224.0.100.18 } vrrp_script chk_down { script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0" interval 3 weight -10 } vrrp_script chk_httpd { script "killall -0 httpd" interval 3 weight -10 } vrrp_instance VI_1 { state BACKUP interface eth0 virtual_router_id 51 priority 98 advert_int 1 authentication { auth_type PASS auth_pass 6Nb6zjWB } virtual_ipaddress { 192.168.11.17 } track_script { chk_down chk_httpd } notify_master "/etc/keepalived/notify.sh master" notify_backup "/etc/keepalived/notify.sh backup" notify_fault "/etc/keepalived/notify.sh fault" } vrrp_instance VI_2 { #相对上面的复制了VI_1内容并改为VI_2,配置也做了相应的修改 state MASTER #由复制BACKUP改为MASTER interface eth0 virtual_router_id 52 #由51改为52 priority 100 #由98改为100 advert_int 1 authentication { auth_type PASS auth_pass 6Nb6zjWB } virtual_ipaddress { 192.168.11.18 #地址192.168.11.17改为192.168.11.18 } track_script { chk_down chk_httpd } notify_master "/etc/keepalived/notify.sh master" notify_backup "/etc/keepalived/notify.sh backup" notify_fault "/etc/keepalived/notify.sh fault" } [[email protected] keepalived]# service keepalived stop #停止keepalived服务 Stopping keepalived: [ OK ] [[email protected] keepalived]# service keepalived start #启动keepalived服务【需要等待两边都stop keepalived 服务，不然会出现一些小问题】 [[email protected] keepalived]# ip add sh #查看VIP地址都在这台，因为另外一台没有开启 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 00:0c:29:2d:ba:b0 brd ff:ff:ff:ff:ff:ff inet 192.168.11.28/24 brd 192.168.11.255 scope global eth0 inet 192.168.11.18/32 scope global eth0 inet 192.168.11.17/32 scope global eth0 inet6 fe80::20c:29ff:fe2d:bab0/64 scope link valid_lft forever preferred_lft forever [[email protected] keepalived]# service keepalived stop #停止keepalived服务 停止 keepalived： [确定] [[email protected] keepalived]# service keepalived start #启动keepalived服务 正在启动 keepalived： [确定] [[email protected] keepalived]# ip add sh #哈哈，我胡三又回来了（192.168.11.17） 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 00:0c:29:ad:98:5c brd ff:ff:ff:ff:ff:ff inet 192.168.11.26/24 brd 192.168.11.255 scope global eth0 inet 192.168.11.17/32 scope global eth0 inet6 fe80::20c:29ff:fead:985c/64 scope link valid_lft forever preferred_lft forever [[email protected] keepalived]# ip add sh #在K2查看，192.168.11.17 没了 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 00:0c:29:2d:ba:b0 brd ff:ff:ff:ff:ff:ff inet 192.168.11.28/24 brd 192.168.11.255 scope global eth0 inet 192.168.11.18/32 scope global eth0 inet6 fe80::20c:29ff:fe2d:bab0/64 scope link valid_lft forever preferred_lft forever You have new mail in /var/spool/mail/root #测试就不测试了 ^-^ 不管是httpd访问停止还是在/etc/keepalived/下touch down 服务都会跑到另外一边去
本文出自 “SunshineBoySZF” 博客，请务必保留此出处http://sunshineboyszf.blog.51cto.com/12087328/1864488
以上是关于Keepalived 的script及双主的主要内容，如果未能解决你的问题，请参考以下文章